Open Access
ARTICLE
An Improved Local RBF Collocation Method for 3D Excavation Deformation Based on Direct Method and Mapping Technique
1 Technology Center, Zhongheng Construction Group Co., Ltd., Nanchang, 330200, China
2 School of Infrastructure Engineering, Nanchang University, Nanchang, 330031, China
3 Zienkiewicz Centre for Computational Engineering, Swansea University, Swansea, SA2 8PP, UK
* Corresponding Author: Hui Zheng. Email:
(This article belongs to the Special Issue: New Trends on Meshless Method and Numerical Analysis)
Computer Modeling in Engineering & Sciences 2025, 142(2), 2147-2172. https://doi.org/10.32604/cmes.2025.059750
Received 16 October 2024; Accepted 27 December 2024; Issue published 27 January 2025
Abstract
Since the plasticity of soil and the irregular shape of the excavation, the efficiency and stability of the traditional local radial basis function (RBF) collocation method (LRBFCM) are inadequate for analyzing three-dimensional (3D) deformation of deep excavation. In this work, the technique known as the direct method, where the local influence nodes are collocated on a straight line, is introduced to optimize the LRBFCM. The direct method can improve the accuracy of the partial derivative, reduce the size effect caused by the large length-width ratio, and weaken the influence of the shape parameters on the LRBFCM. The mapping technique is adopted to transform the physical coordinates of a quadratic-type block to normalized coordinates, in which the deformation problem can easily be solved using the direct method. The stability of the LRBFCM is further modified by considering the irregular shape of 3D excavation, which is divided into several quadratic-type blocks. The soil’s plasticity is described by the Drucker-Prager (D-P) model. The improved LRBFCM is integrated with the incremental method to analyze the plasticity. Five different examples, including strip excavations and circular excavations, are presented to validate the proposed approach’s efficiency.Keywords
Cite This Article

This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.